Profiling Hacker News users based on their comments

2026-03-21 · Source: Simon Willison's Weblog · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Cybersecurity & Data Privacy · Depth: Intermediate, medium

Summary

An experimental method profiles Hacker News users by feeding their last 1,000 comments, obtained via the Algolia API, into a large language model like Claude Opus 4.6. This "startlingly effective" process generates detailed profiles covering professional identity, technical interests, working style, and personality, exemplified by the author's own comprehensive profile. While raising "privacy concerns" due to the depth of derived information, the author primarily uses this tool to identify users who might engage in bad-faith arguments. The LLM can sometimes infer real names, likely from linked personal websites, but this is not consistently observed for all users.

Key takeaway

LLMs like Claude Opus 4.6 can generate "startlingly effective" and highly detailed user profiles from up to 1,000 public Hacker News comments. This process, using Algolia API data, reveals deep insights into professional identity, technical interests, working style, and even personality. While raising significant privacy concerns, this technique offers practical value for understanding user perspectives or identifying bad-faith actors in online moderation.

Topics

• User Profiling
• Large Language Models
• Algolia API
• Agentic Engineering
• Prompt Injection

Code references

• simonw/tools

Best for: AI Engineer, Prompt Engineer, AI Ethicist

Related on AIssential

• Assessing Capabilities of Large Language Models in Social Media Analytics: A Multi-task Quest — Modern LLMs demonstrate varied capabilities across social media analytics tasks, with GPT-4 excelling in authorship verification and Gemini in attribute inference.
• Lexical and Syntactic Characterization of Fake News in Portuguese Produced by Humans and by Machines — Machine-generated Portuguese fake news shows distinct lexical and syntactic patterns compared to human-produced content.
• Companies Are Using Reddit to Manipulate ChatGPT and Google AI Search — Companies are manipulating AI search results and LLM outputs by spamming specific online communities like Reddit.
• MARCA: A Checklist-Based Benchmark for Multilingual Web Search — MARCA evaluates LLM web search reliability using bilingual questions and checklist rubrics.
• Evaluating Reference-Free Summarization Quality Metrics for Portuguese: A Study with Human Judgments in Financial News — Decomposing summarization evaluation into atomic QA pairs surpasses holistic LLM-as-a-Judge methods for Portuguese financial news.
• Subject-level Inference for Realistic Text Anonymization Evaluation — Subject-level inference evaluation is crucial for realistic text anonymization, as span masking alone is insufficient.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Simon Willison's Weblog.